R Notebook

library(tidyverse)
library(tidyr)
library(ggplot2)
library(dplyr)

library(RColorBrewer)

games_sales_full <- read_csv("games_sales_full_cleaned.csv")

Missing column names filled in: 'X1' [1]Parsed with column specification:
cols(
  .default = col_double(),
  name = col_character(),
  genre = col_character(),
  esrb_rating = col_character(),
  platform = col_character(),
  publisher = col_character(),
  developer = col_character(),
  last_update = col_character()
)
See spec(...) for full column specifications.

games_sales_full_total <- read_csv("games_sales_full_total_cleaned.csv")

Missing column names filled in: 'X1' [1]Parsed with column specification:
cols(
  .default = col_double(),
  name = col_character(),
  genre = col_character(),
  esrb_rating = col_character(),
  platform = col_character(),
  publisher = col_character(),
  developer = col_character(),
  last_update = col_character()
)
See spec(...) for full column specifications.

Due to the size of the file I will focuson on the top sales for analysis

games_sales_full_total_top_200 <- games_sales_full_total %>%
  arrange(desc(total_global_sales)) %>%
  slice(1:200)
games_sales_full_total_top_200

games_sales_full_total_top_50 <- games_sales_full_total %>%
  arrange(desc(total_global_sales)) %>%
  slice(1:50)
games_sales_full_total_top_50

summary(games_sales_full_total)

       X1             rank           name              genre           esrb_rating          platform          publisher          developer          critic_score     user_score    total_shipped_2019
 Min.   :    1   Min.   :    1   Length:55792       Length:55792       Length:55792       Length:55792       Length:55792       Length:55792       Min.   : 1.00   Min.   : 2.00   Min.   : 0.0000   
 1st Qu.:13949   1st Qu.:13949   Class :character   Class :character   Class :character   Class :character   Class :character   Class :character   1st Qu.: 6.40   1st Qu.: 7.80   1st Qu.: 0.0000   
 Median :27896   Median :27896   Mode  :character   Mode  :character   Mode  :character   Mode  :character   Mode  :character   Mode  :character   Median : 7.50   Median : 8.50   Median : 0.0000   
 Mean   :27896   Mean   :27896                                                                                                                     Mean   : 7.21   Mean   : 8.25   Mean   : 0.0618   
 3rd Qu.:41844   3rd Qu.:41844                                                                                                                     3rd Qu.: 8.30   3rd Qu.: 9.10   3rd Qu.: 0.0000   
 Max.   :55792   Max.   :55792                                                                                                                     Max.   :10.00   Max.   :10.00   Max.   :82.8600   
                                                                                                                                                   NA's   :49256   NA's   :55457                     
 global_sales_2019 na_sales_2019   eu_sales_2019   jp_sales_2019   other_sales_2019      year      last_update        vgchartzscore   na_sales_2016   eu_sales_2016   jp_sales_2016   other_sales_2016
 Min.   : 0.0000   Min.   :0.00    Min.   :0.00    Min.   :0.00    Min.   :0.00     Min.   :1970   Length:55792       Min.   :2.60    Min.   :0.00    Min.   :0.00    Min.   :0.00    Min.   :0.00    
 1st Qu.: 0.0000   1st Qu.:0.05    1st Qu.:0.01    1st Qu.:0.02    1st Qu.:0.00     1st Qu.:2000   Class :character   1st Qu.:6.80    1st Qu.:0.03    1st Qu.:0.00    1st Qu.:0.00    1st Qu.:0.00    
 Median : 0.0000   Median :0.12    Median :0.04    Median :0.05    Median :0.01     Median :2008   Mode  :character   Median :7.80    Median :0.09    Median :0.01    Median :0.00    Median :0.01    
 Mean   : 0.1272   Mean   :0.28    Mean   :0.16    Mean   :0.11    Mean   :0.04     Mean   :2006                      Mean   :7.43    Mean   :0.14    Mean   :0.06    Mean   :0.01    Mean   :0.02    
 3rd Qu.: 0.0400   3rd Qu.:0.29    3rd Qu.:0.14    3rd Qu.:0.12    3rd Qu.:0.04     3rd Qu.:2011                      3rd Qu.:8.50    3rd Qu.:0.18    3rd Qu.:0.05    3rd Qu.:0.00    3rd Qu.:0.02    
 Max.   :20.3200   Max.   :9.76    Max.   :9.85    Max.   :2.69    Max.   :3.12     Max.   :2020                      Max.   :9.60    Max.   :1.22    Max.   :1.23    Max.   :0.81    Max.   :0.25    
                   NA's   :42828   NA's   :42603   NA's   :48749   NA's   :40270    NA's   :979                       NA's   :54993   NA's   :55238   NA's   :55238   NA's   :55238   NA's   :55238   
 global_sales_2016  total_global_sales
 Min.   :0.000000   Min.   : 0.0000   
 1st Qu.:0.000000   1st Qu.: 0.0000   
 Median :0.000000   Median : 0.0000   
 Mean   :0.002245   Mean   : 0.1912   
 3rd Qu.:0.000000   3rd Qu.: 0.0700   
 Max.   :2.150000   Max.   :82.8600   
                                      

1. Use graph

games_sales_full_total_top_200 %>%
  group_by(genre) %>% 
  ggplot(aes(x = genre, fill = genre) ) +
  geom_bar() +
  labs(title = "Games count by genre") +
  labs(x = "Genre") +
  labs(y = "Number of Games") +
  labs(fill = "Genre") +
  coord_flip()

games_sales_full_total_top_200 %>%
  group_by(genre) %>% 
  ggplot(aes(x = genre, fill = genre) ) +
  geom_bar() +
  theme(legend.position="none") +
  labs(title = "Games count by genre") +
  labs(x = "Genre") +
  labs(y = "Number of Games") +
  labs(fill = "Genre") +
  coord_flip()
ggsave("genre_count.png")

Saving 7.29 x 4.51 in image

2. Use graph ????

games_sales_full_total_top_200 %>% 
  group_by(publisher) %>% 
  ggplot(aes(x = genre, fill = publisher) ) +
  geom_bar() +
  labs(title = "Games count by publisher and genre") +
  labs(x = "Genre") +
  labs(y = "Number of Games") +
  labs(fill = "Publisher") +
  coord_flip()

games_sales_full_total_top_200 %>% 
  group_by(developer) %>% 
  ggplot(aes(x = genre, fill = developer) ) +
  geom_bar() +
  labs(title = "Games count by developer and genre") +
  labs(x = "Genre") +
  labs(y = "Number of Games") +
  labs(fill = "Developer") +
  coord_flip()

games_sales_full_total_top <- games_sales_full_total %>%
  arrange(desc(total_global_sales)) %>%
  slice(1:50)
games_sales_full_total_top

NA

3. Use graph

games_sales_full_total_top %>% 
  group_by(developer) %>% 
  ggplot(aes(x = genre, fill = developer) ) +
  geom_bar() +
  labs(title = "Games count by developer and genre") +
  labs(x = "Genre") +
  labs(y = "Number of Games") +
  labs(fill = "Developer") +
  coord_flip()
ggsave("genre_developer.png")

Saving 7.29 x 4.51 in image

3.5 Use graph - Genre and Dev flipped

games_sales_full_total_top %>% 
  group_by(developer) %>% 
  ggplot(aes(x = developer, fill = genre) ) +
  geom_bar() +
  labs(title = "Games count by developer and genre") +
  labs(x = "Developer") +
  labs(y = "Number of Games") +
  labs(fill = "Genre") +
  coord_flip()
ggsave("developer_genre.png")

Saving 7.29 x 4.51 in image

games_sales_full_total_top_games <- games_sales_full_total %>%
  arrange(desc(total_global_sales)) %>%
  slice(1:20)
games_sales_full_total_top_games

4. Use graph

games_sales_full_total_top_games %>% 
  group_by(name) %>% 
  ggplot(aes(x = name, y = total_global_sales, fill = name)) +
  geom_col() +
  labs(title = "Top 20 games sales by name") +
  labs(x = "Name") +
  labs(y = "Top 20 global sales") +
  labs(fill = "Name") +
  coord_flip()
ggsave("top20_games.png")

Saving 7.29 x 4.51 in image

games_sales_full_total_top_games %>% 
  group_by(name) %>% 
  ggplot(aes(x = name, y = total_global_sales, fill = name)) +
  geom_col() +
  theme(legend.position="none") +
  labs(title = "Top 20 games sales by name") +
  labs(x = "Name") +
  labs(y = "Top 20 global sales") +
  labs(fill = "Name") +
  coord_flip()
ggsave("top20_games.png")

Saving 7.29 x 4.51 in image

games_sales_full_total_top_games_genre <- games_sales_full_total %>%
  arrange(desc(total_global_sales)) %>%
  slice(1:40)
games_sales_full_total_top_games_genre

games_sales_full_total_top_games_genre_select <- games_sales_full_total %>%
  arrange(desc(total_global_sales)) %>%
  select(rank, name, genre, platform, total_global_sales) %>%
  slice(1:50)
games_sales_full_total_top_games_genre_select

games_sales_full_total_top_games_genre %>% 
  group_by(name) %>% 
  ggplot(aes(x = name, y = total_global_sales, fill = genre)) +
  geom_col() +
  labs(title = "Top 40 games sales by name and genre") +
  labs(x = "Name") +
  labs(y = "Top 40 global sales") +
  labs(fill = "Genre") +
  coord_flip()
ggsave("top40_games_name_genre.png")

Saving 7.29 x 4.51 in image

k-means

#games_sales_full_total_top_games_genre %>% 
 # unnest(cols = c(augmented)) %>%
  #filter(k == 2) %>%
# ggplot(aes(x = murder, y = assault, colour = .cluster, label = .rownames)) +
 # geom_point(aes(color = .cluster)) +
 # geom_text(hjust = 0, vjust = - 0.5, size = 3)

5. Use graph

games_sales_full_total_top_200 %>% 
  group_by(developer) %>% 
  ggplot(aes(x = genre, fill = platform)) +
  geom_bar() +
  labs(title = "Games count by platform and genre") +
  labs(x = "Genre") +
  labs(y = "Number of Games") +
  labs(fill = "Platform") +
  coord_flip()
ggsave("count_platform_genre.png")

Saving 7.29 x 4.51 in image

6. Use graph - graph 5. swapped use 5 OR 6

games_sales_full_total_top_200 %>% 
  group_by(developer) %>% 
  ggplot(aes(x = platform, fill = genre)) +
  geom_bar() +
  labs(title = "Games count by genre and platform") +
  labs(x = "Platform") +
  labs(y = "Number of Games") +
  labs(fill = "Genre") +
  coord_flip()
ggsave("count_genre_platform.png")

Saving 7.29 x 4.51 in image

Already used as graph 1

games_sales_full_total_top_200 %>% 
  group_by(genre) %>% 
  ggplot(aes(x = genre, fill = genre)) +
  geom_bar() +
  labs(title = "Games count by genre") +
  labs(x = "Genre") +
  labs(y = "Number of Games") +
  labs(fill = "Genre") +
  coord_flip()
ggsave("genre_count.png")

Saving 7.29 x 4.51 in image

Games by platform

games_sales_full_total_top_200 %>% 
  group_by(genre) %>% 
  ggplot(aes(x = platform, fill = platform) ) +
  geom_bar() +
  labs(title = "Games count by Platform") +
  labs(x = "Platform") +
  labs(y = "Number of Games") +
  labs(fill = "Platform") +
  coord_flip()

7. Use graph

games_sales_full_total_top_200 %>% 
  group_by(genre) %>% 
  ggplot(aes(x = platform, fill = platform) ) +
  geom_bar() +
  theme(legend.position="none") +
  labs(title = "Games count by Platform") +
  labs(x = "Platform") +
  labs(y = "Number of Games") +
  labs(fill = "Platform") +
  coord_flip()
ggsave("platform_count.png")

Saving 7.29 x 4.51 in image

Games by region - global market

games_sales_full_longer %>% 
  group_by(genre) %>% 
  ggplot(aes(x = sales_region, y = sales_region_millions) ) +
  geom_line() +
  labs(title = "Geographic breakdown of Global Sales") +
  labs(x = "Specific global market") +
  labs(y = "Sales (millions)") +
  labs(fill = "Genre") +
  coord_flip()

games_sales_full_total_top_200 %>% 
  group_by(genre) %>% 
  ggplot(aes(x = total_global_sales, fill = genre) ) +
  geom_bar() +
  labs(title = "Geographic breakdown of Global Sales") +
  labs(x = "Sales (millions)") +
  labs(y = "Specific global market") +
  labs(fill = "Genre") +
  coord_flip()

Hide the legend

games_sales_full_total_top_200 %>% 
  group_by(genre) %>% 
  ggplot(aes(x = total_global_sales, fill = genre)) +
  geom_bar() +
  theme(legend.position="none") +
  labs(title = "Geographic breakdown of Global Sales") +
  labs(x = "Sales (millions)") +
  labs(y = "Specific global market") +
  labs(fill = "Genre") +
  coord_flip()

games_sales_full_total_top_200 %>% 
  gather("id", "value", 13:16) %>% 
  ggplot(., aes(total_global_sales, value))+
  geom_point()+
  geom_smooth(method = "lm", se=FALSE, color="black")+
  facet_wrap(~id)

games_sales_full_longer <- read_csv("games_sales_full_longer.csv")

Missing column names filled in: 'X1' [1]Duplicated column names deduplicated: 'X1' => 'X1_1' [2]Parsed with column specification:
cols(
  .default = col_double(),
  name = col_character(),
  genre = col_character(),
  esrb_rating = col_character(),
  platform = col_character(),
  publisher = col_character(),
  developer = col_character(),
  last_update = col_character(),
  na_sales_2016 = col_logical(),
  eu_sales_2016 = col_logical(),
  jp_sales_2016 = col_logical(),
  other_sales_2016 = col_logical(),
  sales_region = col_character()
)
See spec(...) for full column specifications.
6100 parsing failures.
 row              col           expected actual                          file
4441 na_sales_2016    1/0/T/F/TRUE/FALSE   1.21 'games_sales_full_longer.csv'
4441 eu_sales_2016    1/0/T/F/TRUE/FALSE   0.75 'games_sales_full_longer.csv'
4441 other_sales_2016 1/0/T/F/TRUE/FALSE   0.19 'games_sales_full_longer.csv'
4442 na_sales_2016    1/0/T/F/TRUE/FALSE   1.21 'games_sales_full_longer.csv'
4442 eu_sales_2016    1/0/T/F/TRUE/FALSE   0.75 'games_sales_full_longer.csv'
.... ................ .................. ...... .............................
See problems(...) for more details.

games_sales_full_bind_total_longer <- read_csv("games_sales_full_bind_total_longer.csv")

Missing column names filled in: 'X1' [1]Parsed with column specification:
cols(
  X1 = col_double(),
  name = col_character(),
  platform = col_character(),
  genre = col_character(),
  publisher = col_character(),
  global_sales = col_double(),
  critic_score = col_double(),
  developer = col_character(),
  year_data = col_double(),
  rank = col_logical(),
  esrb_rating = col_logical(),
  user_score = col_logical(),
  total_shipped = col_logical(),
  year = col_logical(),
  last_update = col_logical(),
  vgchartzscore = col_logical(),
  total_global_sales = col_logical(),
  sales_region = col_character(),
  sales_region_millions = col_double()
)
560728 parsing failures.
  row           col           expected actual                                     file
66877 esrb_rating   1/0/T/F/TRUE/FALSE  E     'games_sales_full_bind_total_longer.csv'
66877 total_shipped 1/0/T/F/TRUE/FALSE  82.86 'games_sales_full_bind_total_longer.csv'
66877 year          1/0/T/F/TRUE/FALSE  2006  'games_sales_full_bind_total_longer.csv'
66878 esrb_rating   1/0/T/F/TRUE/FALSE  E     'games_sales_full_bind_total_longer.csv'
66878 total_shipped 1/0/T/F/TRUE/FALSE  82.86 'games_sales_full_bind_total_longer.csv'
..... ............. .................. ...... ........................................
See problems(...) for more details.

games_sales_full_bind_total_longer %>%
  group_by(sales_region, year_data) %>%
  summarise(Total = sum(sales_region_millions, na.rm = TRUE))

`summarise()` regrouping output by 'sales_region' (override with `.groups` argument)

games_sales_full_bind_total_longer %>%
  group_by(sales_region, year_data) %>%
  summarise(Total = sum(sales_region_millions, na.rm = TRUE)) %>%
  ggplot(aes(x = sales_region, y = Total)) +
  geom_bar(stat = "identity") +
  labs(title = "Total gobal sales by region") +
  labs(x = "Global market region") +
  labs(y = "Total sales (millions)") +
  facet_wrap(~year_data)

`summarise()` regrouping output by 'sales_region' (override with `.groups` argument)

ggplot(games_sales_full_bind_total_longer, aes(sales_region, sales_region_millions)) +
  geom_boxplot() +
  labs(title = "Distribution of sales by genre") +
  labs(x = "Genre") +
  labs(y = "Sales (millions)") +
   coord_flip()

8. Use graph

games_sales_full_bind_total_longer_region %>%
  ggplot(aes(x = sales_region, y = sales_region_millions)) +
  geom_col() +
  labs(title = "Total gobal sales by region") +
  labs(x = "Global market region") +
  labs(y = "Total sales (millions)") +
  scale_fill_brewer () +
  facet_wrap(~year_data)

games_sales_full_bind_total<- read_csv("games_sales_full_bind_total.csv")

Missing column names filled in: 'X1' [1]Parsed with column specification:
cols(
  .default = col_double(),
  name = col_character(),
  platform = col_character(),
  genre = col_character(),
  publisher = col_character(),
  developer = col_character(),
  rank = col_logical(),
  esrb_rating = col_logical(),
  user_score = col_logical(),
  total_shipped = col_logical(),
  year = col_logical(),
  last_update = col_logical(),
  vgchartzscore = col_logical(),
  total_global_sales = col_logical()
)
See spec(...) for full column specifications.
140182 parsing failures.
  row           col           expected actual                              file
16720 esrb_rating   1/0/T/F/TRUE/FALSE  E     'games_sales_full_bind_total.csv'
16720 total_shipped 1/0/T/F/TRUE/FALSE  82.86 'games_sales_full_bind_total.csv'
16720 year          1/0/T/F/TRUE/FALSE  2006  'games_sales_full_bind_total.csv'
16721 rank          1/0/T/F/TRUE/FALSE  2     'games_sales_full_bind_total.csv'
16721 total_shipped 1/0/T/F/TRUE/FALSE  40.24 'games_sales_full_bind_total.csv'
..... ............. .................. ...... .................................
See problems(...) for more details.

Dont use

games_sales_full_bind_total_longer %>%
  ggplot(aes(x = sales_region, y = sales_region_millions, fill = genre)) +
  geom_col() +
  labs(title = "Total global sales by region") +
  labs(x = "Total sales") +
  labs(y = "Global market region")
ggsave("sales_region_genre.png")

Saving 7.29 x 4.51 in image

XXXXXXXX WRONG DS

games_sales_full_bind_total_longer %>%
  ggplot(aes(x = genre, y = sales_region_millions, fill = sales_region)) +
  geom_col() +
  labs(title = "Total global sales by genre and region") +
  labs(x = "Global market region") +
  labs(y = "Total sales") +
  coord_flip()
ggsave("sales_genre_region.png")

Saving 7.29 x 4.51 in image

Correct DS

games_sales_full_bind_total_longer %>%
  group_by(sales_region, year_data, genre, sales_region_millions) %>%
  summarise(Total = sum(sales_region_millions, na.rm = TRUE)) %>%
  ggplot(aes(x = genre, y = sales_region_millions, fill = sales_region)) +
  geom_col() +
  labs(title = "Total global sales by genre and region") +
  labs(x = "Global market region") +
  labs(y = "Total sales") +
  coord_flip()

`summarise()` regrouping output by 'sales_region', 'year_data', 'genre' (override with `.groups` argument)

ggsave("sales_genre_region.png")

Saving 7.29 x 4.51 in image

9. Use graph

Games mean (max/ min) sales

ggplot(games_sales_full_total_top_200, aes(genre, total_global_sales)) +
  geom_boxplot() +
  labs(title = "Distribution of sales by genre") +
  labs(x = "Genre") +
  labs(y = "Sales (millions)") +
   coord_flip()
ggsave("distribution_sales_by_genre.png")

Saving 7.29 x 4.51 in image

10. Use graph

ggplot(games_sales_full_total_top_200, aes(platform, total_global_sales)) +
  geom_boxplot() +
  labs(title = "Distribution of sales by platform") +
  labs(x = "Platform") +
  labs(y = "Sales (millions)") +
   coord_flip()
ggsave("distribution_sales_by_platform.png")

Saving 7.29 x 4.51 in image

shipped / critic score